Iris Nebula and PixInsight - combining different exposures

[BrendanC]

Hi all,

I've been gathering data for the Iris nebula this month. I want to do as good a job as I possibly can on it, and so I've decided that I'm going to collect three sets: one for the overall image at 120s subs (this is with an ASI533MC Pro and 130PDS, that sub length works fine for noise levels); one at 30s exposures to try and get as much detail from the core as possible which is often blown out; and 300s so that I can get as much of the surrounding dust as possible.

The idea is to stack them separately, remove the stars, process, layer and mask in Photoshop, then put the stars from the 120s group back in.

Or, would it be best just to stack the whole lot in one go?

Or, should I just not do this at all and go for the standard 120s, and assume the core will be fine, and that I'll be able to process the dust without additional ultra-long exposures?

Any/all advice appreciated!

Thanks
Brendan

[Budgie1]

If you're using WBPP in PixInsight for stacking then I would do both. Stack the whole lot together to see what you get, then stack them separately and combine them using ImageIntegration in PixInsight. 

You can then compare the resulting images. You will get the combined stars in the whole stack but you can remove them from the separately stacked images before combination and then add just the 120s set back in after.

If you've gone to the trouble of shooting the 30s & 300s sets, it's a shame not to use them.

[BrendanC]

On it like a car bonnet. 

Thanks! 

[Fegato]

I think the core can blow out due to the stretching, as you're often trying to bring out all the dust and dark stuff in this area. So I've tackled it with masks (range masks in particular), having used my "standard" exposure length for broadband (30s in my case). If it feels too bright even after that, you can try HDRMultiscaleTransform - I've sometimes had success with this. Or you could even use full HDR processing if the core really is blown out by 300s and you have the multiple exposure times. It's a tricky one, as the more you stretch the dust, the more you have to bring up the core anyway in order to keep it looking "natural", there's a limit to how much you can sensibly mask.

Here's my latest attempt at this - https://astrob.in/fw81wa/0/ - I've really brought the dust out, so the core is quite bright

[BrendanC]

"My God, it's full of dust."

Awesome image.

I've come across HDRMultiscaleTransform before and intended to find out more about it, perhaps now is the time.

Also, I'm intrigued by standard 30s exposures! I guess that with an 11" RASA you have a huge light bucket which enables that. I'm going for 30s for the core but I'm generally about right with broadband at 120s gain 101 with my rig and conditions.

After previewing the data last night I'm going to ditch the 300s stuff because it's just overexposing everything and not really improving things, so the next couple of nights it's going to be 120s as per usual, and I might just try an hour or so of 15s to really get to the heart of the nebula, just as an experiment. That's what I did with my M42 and captured the trapezium, and how I approached my M3 too recently to avoid it just being a big bright blob. Ended up looking like a dandelion!

 

Thanks for the advice all, really appreciate it.

Edited May 19, 2023 by BrendanC

[BrendanC]

This is probably about right!

 

[ollypenrice]

Before deviating from your standard sub length I would simply use your regular length and see if the core is over exposed at the linear stage. If it isn't, you don't need short subs, you just need a masked or layered stretch. (I'd use PS layer masks rather the very Pixinsighty-looking HDR wavelets.)  I haven't done this target with a CMOS but I resolved the core star completely in 15 minute luminance subs with a CCD. It really is only that star which is likely to be a problem. Just away from it you need deep data to catch the faint but tantalizing pinks which live there. If you do need shorts for the star you only need a few subs because, so close to saturation, there will be no noise to worry about.

In my view 120 sec and 300 sec will give you essentially the same result except that one will be slightly better than the other. I would just go with whichever is best because the better one will be better across the board.  I also think, based on Trapezium experience, that 30 seconds will be too long to be a big help on the central star. I'd just shoot half a dozen 10 second subs to rescue the star if needed.

Olly

Edit: I found an old Iris core done with full length CCD exposures. There are things I could do much better now but it makes the case for the longer subs. Note the faint pinks right next to the star. Those need signal.

Frankly, I didn't like the PI result above very much.

Edited May 20, 2023 by ollypenrice

[BrendanC]

Outstanding. Thank you Olly.